Defect Detection Research Articles

YOLO-DHGC: Small Object Detection Using Two-Stream Structure with Dense Connections.

Small object detection, which is frequently applied in defect detection, medical imaging, and security surveillance, often suffers from low accuracy due to limited feature information and blurred details. This paper proposes a small object detection method named YOLO-DHGC, which employs a two-stream structure with dense connections. Firstly, a novel backbone network, DenseHRNet, is introduced. It innovatively combines a dense connection mechanism with high-resolution feature map branches, effectively enhancing feature reuse and cross-layer fusion, thereby obtaining high-level semantic information from the image. Secondly, a two-stream structure based on an edge-gated branch is designed. It uses higher-level information from the regular detection stream to eliminate irrelevant interference remaining in the early processing stages of the edge-gated stream, allowing it to focus on processing information related to shape boundaries and accurately capture the morphological features of small objects. To assess the effectiveness of the proposed YOLO-DHGC method, we conducted experiments on several public datasets and a self-constructed dataset. Exceptionally, a defect detection accuracy of 96.3% was achieved on the Market-PCB public dataset, demonstrating the effectiveness of our method in detecting small object defects for industrial applications.

Data enhanced YOLOv8s algorithm for X-ray weld defect detection

Data enhanced YOLOv8s algorithm for X-ray weld defect detection

Experimental study on infrared detection of debonding in concrete-filled steel tubular structure under acceleratory period of hydration heat action

Concrete-filled steel tubular (CFST) Structures often face challenges with debonding during construction, which can markedly compromise the structural integrity. The hydration of concrete can generate significant heat during construction, making the infrared thermography as a potential method for the defect detection. However, effects of concrete hydration behavior, debonding size, and environmental factors on the infrared thermal imaging of CFST debonding remains unclear. This study conducted experiments using four different hydration heating rates and three debonding sizes to simulate debonding detection using infrared imaging during the hydration phase of CFST. The feasibility of the simulation approach was validated through pair-to-pair comparison, and multiple linear regression analysis was employed to evaluate the impact of these factors. The findings highlight that absolute temperature difference has the most significant impact on detection effectiveness regardless of interaction effects. In regression models without interaction, heating rate demonstrated the least impact, Whereas the model considering the interaction showed that the interaction effect of heating rate and debonding size showed a secondary effect. Further examination indicated that interaction effects decrease as heating rate and debonding size decrease.

Development of automation and monitoring methods for narrow-gap hot-wire laser welding using camera images

AbstractIn this study, an image-based method was developed for hot-wire laser narrow gap welding. The welding process was monitored based on image information processed using semantic segmentation, a method of classifying images by pixel. To control the welding position, an experimental system was configured that automatically follows the welding position by recognizing the position of the welding groove from the image during welding. In monitoring weld defects, a method was developed to predict the lack of fusion occurring on the wall surface using brightness information near the wall surface. For the lack of fusion occurring at the bottom of the groove, a defect detection method was developed by monitoring the molten pool shape using semantic segmentation. Defects were generated by intentionally reducing the laser power, and the defects were monitored from images taken during processing. In the unstable state where the laser power was reduced, the shape in front of the molten pool became unstable, and the occurrence of defects was monitored by capturing the shape change. In conclusion, this research made it possible to control and monitor the welding process with a single camera.

PDeT: A Progressive Deformable Transformer for Photovoltaic Panel Defect Segmentation.

Defects in photovoltaic (PV) panels can significantly reduce the power generation efficiency of the system and may cause localized overheating due to uneven current distribution. Therefore, adopting precise pixel-level defect detection, i.e., defect segmentation, technology is essential to ensuring stable operation. However, for effective defect segmentation, the feature extractor must adaptively determine the appropriate scale or receptive field for accurate defect localization, while the decoder must seamlessly fuse coarse-level semantics with fine-grained features to enhance high-level representations. In this paper, we propose a Progressive Deformable Transformer (PDeT) for defect segmentation in PV cells. This approach effectively learns spatial sampling offsets and refines features progressively through coarse-level semantic attention. Specifically, the network adaptively captures spatial offset positions and computes self-attention, expanding the model's receptive field and enabling feature extraction across objects of various shapes. Furthermore, we introduce a semantic aggregation module to refine semantic information, converting the fused feature map into a scale space and balancing contextual information. Extensive experiments demonstrate the effectiveness of our method, achieving an mIoU of 88.41% on our solar cell dataset, outperforming other methods. Additionally, to validate the PDeT's applicability across different domains, we trained and tested it on the MVTec-AD dataset. The experimental results demonstrate that the PDeT exhibits excellent recognition performance in various other scenarios as well.

A Classification and Segmentation Model for Diamond Abrasive Grains Based on Improved Swin-Unet-SAM

The detection of abrasive grain images in diamond tools serves as the foundation for assessing the overall condition of the tools, encompassing crucial aspects of diamond abrasive grains like the quantity, size, morphology, and distribution. Given the intricate background textures and reflective characteristics exhibited by diamond images, diamond detection and segmentation pose a significant challenge. Recently, numerous defect detection methods based on machine learning and deep learning have emerged. However, several issues persist, such as detection accuracy and the interference caused by intricate background textures. The present work demonstrates an efficient classification and segmentation network algorithm that combines Swin-Unet with SAM (Segment Anything Model) to alleviate the existing problems. Specifically, four embedding structures were devised to bridge the two models for iterative training. The transformer blocks within the Swin-Unet model were enhanced to facilitate classification and coarse segmentation, and the mask structure in SAM was refined to enable fine segmentation. The experimental results show that under a small sample dataset with complex background textures, the average index values of ACC (accuracy), SE (Sensitivity), and DSC (Dice Similarity Coefficient) for the classification and segmentation of diamond abrasive grains reached 98.7%, 92.5%, and 85.9%, respectively. Compared with the model before improvement, its ACC, SE and DSC increased by 1.2%, 15.9%, and 7.6%, respectively. The test results, based on four different datasets, consistently indicated that this model has excellent segmentation performance and robustness and has great application potential in the industrial field.

Unmanned aerial vehicle transmission defect detection technology based on edge computing

Unmanned aerial vehicle transmission defect detection technology based on edge computing

Hyper CLS-Data-Based Robotic Interface and Its Application to Intelligent Peg-in-Hole Task Robot Incorporating a CNN Model for Defect Detection

Various types of numerical control (NC) machine tools can be standardly operated and controlled based on NC data that can be easily generated using widespread CAD/CAM systems. On the other hand, the operation environments of industrial robots still depend on conventional teaching and playback systems provided by the makers, so it seems that they have not been standardized and unified like NC machine tools yet. Additionally, robotic functional extensions, e.g., the easy implementation of a machine learning model, such as a convolutional neural network (CNN), a visual feedback controller, cooperative control for multiple robots, and so on, has not been sufficiently realized yet. In this paper, a hyper cutter location source (HCLS)-data-based robotic interface is proposed to cope with the issues. Due to the HCLS-data-based robot interface, the robotic control sequence can be visually and unifiedly described as NC codes. In addition, a VGG19-based CNN model for defect detection, whose classification accuracy is over 99% and average time for forward calculation is 70 ms, can be systematically incorporated into a robotic control application that handles multiple robots. The effectiveness and validity of the proposed system are demonstrated through a cooperative pick and place task using three small-sized industrial robot MG400s and a peg-in-hole task while checking undesirable defects in workpieces with a CNN model without using any programmable logic controller (PLC). The specifications of the PC used for the experiments are CPU: Intel(R) Core(TM) i9-10850K CPU 3.60 GHz, GPU: NVIDIA GeForce RTX 3090, Main memory: 64 GB.

Improved Real-Time Detection Transformer-Based Rail Fastener Defect Detection Algorithm

To address the issues of the Real-Time DEtection TRansformer (RT-DETR) object detection model, including poor defect feature extraction in the task of rail fastener defect detection, inefficient use of computational resources, and suboptimal channel attention in the self-attention mechanism, the following improvements were made. Firstly, a Super-Resolution Convolutional Module (SRConv) was designed as a separate component and integrated into the Backbone network, which enhances the image details and clarity while preserving the original image structure and semantic content. This integration improves the model’s ability to extract defect features. Secondly, a channel attention mechanism was integrated into the self-attention module of RT-DETR to enhance the focus on feature map channels, addressing the problem of sparse attention maps caused by the lack of channel attention while saving computational resources. Finally, the experimental results show that compared to the original model, the improved RT-DETR-based rail fastener defect detection algorithm, with an additional 0.4 MB of parameters, achieved a higher accuracy, with a 2.8 percentage point increase in the Mean Average Precision (mAP) across IoU thresholds from 0.5 to 0.9 and a 1.7 percentage point increase in the Average Recall (AR) across the same thresholds.

Research on tire internal defect identification method based on deep learning

As an important part of automobile, the safety and durability of tire have attracted more and more attention. Tire defect detection is an important link to ensure tire quality, while traditional detection methods have problems such as low efficiency, high false detection rate and high labor intensity. Therefore, this study aims to develop an efficient and accurate tire defect identification and classification technique to improve the efficiency and accuracy of tire inspection. In this paper, based on YOLO (You Only Look Once) v5 algorithm, tire defect recognition and classification are studied. Firstly, the data sets containing various types of tire defects were collected and sorted out, and the data sets were preprocessed. Then, by constructing, training and optimizing the YOLOv5 tire defect recognition model, the fast and accurate recognition of tire defects was realized. Finally, the performance of the model was evaluated through experiments and compared with the existing methods. The experimental results show that the tire defect recognition and classification method based on YOLOv5 proposed in this study has high accuracy. Compared with traditional methods, this method has a significant improvement in detection speed and accuracy.

Stress and defect detection of specimens based on tag array sensing technology

Stress and defect detection of specimens based on tag array sensing technology

Flaw detection of railway catenary insulator based on DP-YOLOv5 algorithm with bright and dark channel enhancement

Abstract In order to ensure the timely detection of safety hazards in overhead transmission lines with railroad conductors and improve the accuracy of night insulator defect detection, this paper proposes the DP-YOLOv5 algorithm with dark and light channel enhancement optimization. It improves the night insulator image quality by introducing the dark and light channel enhancement algorithm, builds a lightweight network by combining the DP-BS module, and adds the Shuffle Attention module to enhance the feature extraction and ensure detection accuracy. At the same time, the EC-Loss loss function is used to optimize the prediction frame adjustment, accelerate the model convergence, and improve detection efficiency and accuracy. The simulation results show that the accuracy of DP-YOLOv5 is 95.3%, the recall is 94.8%, the average accuracy is 95.5%, and the FLOPs are 219.3. Compared with YOLOv5, the mapped value is improved by 0.9%, the F1 is improved by 1%, and the model parameter and FLOPs are reduced by 48.8% and 50.8%, respectively.

Detecting Multi-Scale Defects in Material Extrusion Additive Manufacturing of Fiber-Reinforced Thermoplastic Composites: A Review of Challenges and Advanced Non-Destructive Testing Techniques.

Additive manufacturing (AM) defects present significant challenges in fiber-reinforced thermoplastic composites (FRTPCs), directly impacting both their structural and non-structural performance. In structures produced through material extrusion-based AM, specifically fused filament fabrication (FFF), the layer-by-layer deposition can introduce defects such as porosity (up to 10-15% in some cases), delamination, voids, fiber misalignment, and incomplete fusion between layers. These defects compromise mechanical properties, leading to reduction of up to 30% in tensile strength and, in some cases, up to 20% in fatigue life, severely diminishing the composite's overall performance and structural integrity. Conventional non-destructive testing (NDT) techniques often struggle to detect such multi-scale defects efficiently, especially when resolution, penetration depth, or material heterogeneity pose challenges. This review critically examines manufacturing defects in FRTPCs, classifying FFF-induced defects based on morphology, location, and size. Advanced NDT techniques, such as micro-computed tomography (micro-CT), which is capable of detecting voids smaller than 10 µm, and structural health monitoring (SHM) systems integrated with self-sensing fibers, are discussed. The role of machine-learning (ML) algorithms in enhancing the sensitivity and reliability of NDT methods is also highlighted, showing that ML integration can improve defect detection by up to 25-30% compared to traditional NDT techniques. Finally, the potential of self-reporting FRTPCs, equipped with continuous fibers for real-time defect detection and in situ SHM, is investigated. By integrating ML-enhanced NDT with self-reporting FRTPCs, the accuracy and efficiency of defect detection can be significantly improved, fostering broader adoption of AM in aerospace applications by enabling the production of more reliable, defect-minimized FRTPC components.

Online Quality Control of Powder Bed Fusion with High-Resolution Eddy Current Testing Inductive Sensor Arrays.

This paper presents the development of a novel eddy current array (ECA) system for real-time, layer-by-layer quality control in powder bed fusion (PBF) additive manufacturing. The system is integrated into the recoater of a PBF machine to provide spatially resolved electrical conductivity imaging of the manufactured part. The system features an array of 40 inductive sensors spaced at 1 mm pitch and is capable of performing a full array readout every 0.192 mm at 100 mm/s recoater speed. Array scalability was achieved through the careful selection of the electromagnetic configuration, miniaturized and seamlessly integrated sensor elements, and the use of advanced mixed signal processing techniques. Experimental validation was performed on stainless steel 316L parts, successfully detecting metallic structures and confirming system performance in both laboratory and real-time PBF environments. The prototype achieved a signal-to-noise ratio (SNR) of 26.5 dB, discriminating metal from air and thus demonstrating its potential for improving PBF part design, process optimization, and defect detection.

LWFDD-YOLO: a lightweight defect detection algorithm based on improved YOLOv8

LWFDD-YOLO: a lightweight defect detection algorithm based on improved YOLOv8

Steel Surface Defect Detection Based on YOLOv8-TLC

Steel Surface Defect Detection Based on YOLOv8-TLC

Low-dose radiographic inspection of welding by a novel aperiodic reverse stochastic resonance method

Abstract Low-dose radiographic inspection is a growing trend in industry to minimize radiation risks to humans and the environment. However, reduction in radiation dose often introduces significant noise, which affects image quality and hinders accurate identification of subtle defects. This study addresses this issue by introducing a novel phenomenon called aperiodic reverse stochastic resonance (ARSR), observed in nonlinear systems excited by aperiodic binary signals. ARSR enables simultaneous amplitude amplification and reversal of signals under specific noise conditions. Leveraging ARSR, we propose an image denoising framework for low-dose radiographic inspections. First, a set of projection data is obtained by using Radon transform to reduce the dimensionality of X-ray images from different angles. Then, the projection data is modulated based on the ARSR system. Finally, the image is reconstructed based on the inverse Radon transform. Simulations and experimental comparison results in welding applications validate the effectiveness of the framework, demonstrating significant improvements in image quality for low-dose radiographic defect detection. Unlike advanced methods such as Gaussian filtering, BM3D, and DnCNN, which operate at the pixel level, ARSR performs denoising at the projection data stage, reducing noise impact, preserving original information, and focusing on physical data processing during imaging. This approach enhances the detection of subtle defects, highlighting the potential of stochastic resonance in image processing.

A novel wafer defocus measurement method for spot-scanning imaging system using laser triangulation

Abstract For spot scanning dark-field scattering technology, the defocus caused by the change in wafer surface height decreases the defect detection rate and size measurement repeatability. The demand for accurately and rapidly measuring the wafer surface height online is becoming increasingly urgent. The defect detection system integrates scattering, reflection, film thickness, and topography defect detection channels. The integration of laser-triangulation-based defocus measurement into the bright-field optical path is proposed. The laser beam is directed onto the surface of the wafer, and the reflected light is transmitted through a lens onto the photosensitive surface of a four-quadrant detector. This detector captures both the strength and position of the reflected signal simultaneously. In this paper, a triangulation-modified model is established to obtain the height of each inspection spot on the wafer surface through the position signal of the reflected image. When the signal-to-noise ratio of the simulated reflection image position is 20 dB, the signal-to-noise ratio of the height measurement resolved by the triangulation-modified model is 41 dB, and the measurement error is less than 0.2 μm, indicating that the model has a noise suppression effect. The spot-scanning imaging system is established, and the height measurement results of the triangulation measurement model are verified using a laser interferometer. Within the travel range of 69 μm, the system has a measurement accuracy better than ±1 μm and a measurement repeatability better than ±0.2 μm, which meet the measurement requirements of the spot-scanning imaging system. The system is utilized for defect detection and defocus measurement of the wafer surface, along with the motion mechanism, to accurately and rapidly obtain the height distribution of the wafer surface.

Defect Detection of Polyethylene Gas Pipeline Based on Convolutional Neural Networks and Image Processing

Abstract In this paper, a method based on image recognition was proposed to detect the defects of polyethylene (PE) gas pipeline, especially the deformation due to the indentation. First, the pipeline -detection VGG (PD-VGG) model was established based on the convolutional neural network (CNN), and appropriate model parameters were optimized through model training. The defect recognition rate of the improved model can reach 94.76%. Following, the weighted average graying algorithm was used to separate the defects characterized by deformation. Then, an improved gamma correction algorithm was applied to achieve image enhancement, and the interference of impurities adhered on intersurface of pipeline was also removed by using multilayer filters. The edge detection of the defect image was completed by using the Canny operator, and following the screening between the target contour and the interference contour by using top-contour. Finally, the algorithm for minimum outer rectangle algorithm was used to fit the defect contour, and the eigenvalues of deformation defects were extracted. The results indicate that the above defect detection method can better extract the deformation contour of the dented pipeline. The high agreement with the experimental results provides a basis for the research of effectively recognizing whether the pipeline has undergone ductile failure only through profile detection of defects.

Road Defect Identification and Location Method Based on an Improved ML-YOLO Algorithm.

The conventional method for detecting road defects relies heavily on manual inspections, which are often inefficient and struggle with precise defect localization. This paper introduces a novel approach for identifying and locating road defects based on an enhanced ML-YOLO algorithm. By refining the YOLOv8 object detection framework, we optimize both the traditional convolutional layers and the spatial pyramid pooling network. Additionally, we incorporate the Convolutional Block Attention to effectively capture channel and spatial features, along with the Selective Kernel Networks that dynamically adapt to feature extraction across varying scales. An optimized target localization algorithm is proposed to achieve high-precision identification and accurate positioning of road defects. Experimental results indicate that the detection accuracy of the improved ML-YOLO algorithm reaches 0.841, with a recall rate of 0.745 and an average precision of 0.817. Compared to the baseline YOLOv8 model, there is an increase in accuracy by 0.13, a rise in recall rate by 0.117, and an enhancement in average precision by 0.116. After the high detection accuracy of road defects was confirmed, generalization experiments were carried out on the improved ML-YOLO model in the public data set. The experimental results showed that compared with the original YOLOv8n, the average precision and recall rate of all types of ML-YOLO increased by 0.075, 0.121, and 0.035 respectively, indicating robust generalization capabilities. When applied to real-time road monitoring scenarios, this algorithm facilitates precise detection and localization of defects while significantly mitigating traffic accident risks and extending roadway service life. A high detection accuracy of road defects was achieved.